minimap2 0.2.26.1 → 0.2.28.0

data/ext/minimap2/minimap2.1

@@ -1,4 +1,4 @@
-.TH minimap2 1 "29 April 2023" "minimap2-2.26 (r1175)" "Bioinformatics tools"
+.TH minimap2 1 "12 March 2024" "minimap2-2.28 (r1209)" "Bioinformatics tools"
 .SH NAME
 .PP
 minimap2 - mapping and alignment between collections of DNA sequences
@@ -268,6 +268,11 @@ or more of the shorter chain [0.5]
 Use the minigraph chaining algorithm [no]. The minigraph algorithm is better
 for aligning contigs through long INDELs.
 .TP
+.BI --rmq-inner \ NUM
+Apply full dynamic programming for anchors within distance
+.I NUM
+[1000].
+.TP
 .B --hard-mask-level
 Honor option
 .B -M
@@ -343,6 +348,10 @@ Matching score [2]
 .BI -B \ INT
 Mismatching penalty [4]
 .TP
+.BI -b \ INT
+Mismatching penalty for transitions [same as
+.BR -B ].
+.TP
 .BI -O \ INT1[,INT2]
 Gap open penalty [4,24]. If
 .I INT2
@@ -356,10 +365,19 @@ costs
 .RI min{ O1 + k * E1 , O2 + k * E2 }.
 In the splice mode, the second gap penalties are not used.
 .TP
+.BI -J \ INT
+Splice model [1]. 0 for the original minimap2 splice model that always penalizes non-GT-AG splicing;
+1 for the miniprot model that considers non-GT-AG. Option
+.B -C
+has no effect with the default
+.BR -J1 .
+.BR -J0 .
+.TP
 .BI -C \ INT
 Cost for a non-canonical GT-AG splicing (effective with
-.BR --splice )
-[0]
+.B --splice
+.BR -J0 )
+[0].
 .TP
 .BI -z \ INT1[,INT2]
 Truncate an alignment if the running alignment score drops too quickly along
@@ -450,7 +468,7 @@ Set 0 to disable [100m].
 .BI --cap-kalloc \ NUM
 Free thread-local kalloc memory reservoir if after the alignment the size of the reservoir above
 .IR NUM .
-Set 0 to disable [0].
+Set 0 to disable [500m].
 .SS Input/output options
 .TP 10
 .B -a
@@ -506,6 +524,9 @@ Output =/X CIGAR operators for sequence match/mismatch.
 .B -Y
 In SAM output, use soft clipping for supplementary alignments.
 .TP
+.B --secondary-seq
+In SAM output, show query sequences for secondary alignments.
+.TP
 .BI --seed \ INT
 Integer seed for randomizing equally best hits. Minimap2 hashes
 .I INT
@@ -566,15 +587,43 @@ are:
 Align noisy long reads of ~10% error rate to a reference genome. This is the
 default mode.
 .TP
+.B lr:hq
+Align accurate long reads (error rate <1%) to a reference genome
+.RB ( -k19
+.B -w19 -U50,500
+.BR -g10k ).
+This was recommended by ONT developers for recent Nanopore reads
+produced with chemistry v14 that can reach ~99% in accuracy.
+It was shown to work better for accurate Nanopore reads
+than
+.BR map-hifi .
+.TP
 .B map-hifi
 Align PacBio high-fidelity (HiFi) reads to a reference genome
-.RB ( -k19
-.B -w19 -U50,500 -g10k -A1 -B4 -O6,26 -E2,1
+.RB ( -xlr:hq
+.B -A1 -B4 -O6,26 -E2,1
 .BR -s200 ).
+It differs from
+.B lr:hq
+only in scoring. It has not been tested whether
+.B lr:hq
+would work better for PacBio HiFi reads.
 .TP
 .B map-pb
 Align older PacBio continuous long (CLR) reads to a reference genome
 .RB ( -Hk19 ).
+Note that this data type is effectively deprecated by HiFi.
+Unless you work on very old data, you probably want to use
+.B map-hifi
+or
+.BR lr:hq .
+.TP
+.B map-iclr
+Align Illumina Complete Long Reads (ICLR) to a reference genome
+.RB ( -k19
+.B -B6 -b4
+.BR -O10,50 ).
+This was recommended by Illumina developers.
 .TP
 .B asm5
 Long assembly to reference mapping
@@ -582,21 +631,21 @@ Long assembly to reference mapping
 .B -w19 -U50,500 --rmq -r1k,100k -g10k -A1 -B19 -O39,81 -E3,1 -s200 -z200
 .BR -N50 ).
 Typically, the alignment will not extend to regions with 5% or higher sequence
-divergence. Only use this preset if the average divergence is far below 5%.
+divergence. Use this preset if the average divergence is not much higher than 0.1%.
 .TP
 .B asm10
 Long assembly to reference mapping
 .RB ( -k19
 .B -w19 -U50,500 --rmq -r1k,100k -g10k -A1 -B9 -O16,41 -E2,1 -s200 -z200
 .BR -N50 ).
-Up to 10% sequence divergence.
+Use this if the average divergence is around 1%.
 .TP
 .B asm20
 Long assembly to reference mapping
 .RB ( -k19
 .B -w10 -U50,500 --rmq -r1k,100k -g10k -A1 -B4 -O6,26 -E2,1 -s200 -z200
 .BR -N50 ).
-Up to 20% sequence divergence.
+Use this if the average divergence is around several percent.
 .TP
 .B splice
 Long-read spliced alignment
@@ -612,13 +661,13 @@ costs are different during chaining; 4) the computation of the
 tag ignores introns to demote hits to pseudogenes.
 .TP
 .B splice:hq
-Long-read splice alignment for PacBio CCS reads
+Spliced alignment for accurate long RNA-seq reads such as PacBio iso-seq
 .RB ( -xsplice
 .B -C5 -O6,24
 .BR -B4 ).
 .TP
 .B sr
-Short single-end reads without splicing
+Short-read alignment without splicing
 .RB ( -k21
 .B -w11 --sr --frag=yes -A2 -B8 -O12,32 -E2,1 -b0 -r100 -p.5 -N20 -f1000,5000 -n2 -m25
 .B -s40 -g100 -2K50m --heap-sort=yes

data/ext/minimap2/misc/paftools.js

@@ -1,6 +1,6 @@
 #!/usr/bin/env k8
 
-var paftools_version = '2.26-r1175';
+var paftools_version = '2.28-r1209';
 
 /*****************************
  ***** Library functions *****
@@ -133,26 +133,50 @@ Interval.find_ovlp = function(a, st, en)
 
 function fasta_read(fn)
 {
-	var h = {}, gt = '>'.charCodeAt(0);
+	var h = {}, seqlen = [];
+	var buf = new Bytes();
 	var file = fn == '-'? new File() : new File(fn);
-	var buf = new Bytes(), seq = null, name = null, seqlen = [];
-	while (file.readline(buf) >= 0) {
-		if (buf[0] == gt) {
-			if (seq != null && name != null) {
-				seqlen.push([name, seq.length]);
-				h[name] = seq;
-				name = seq = null;
-			}
-			var m, line = buf.toString();
-			if ((m = /^>(\S+)/.exec(line)) != null) {
-				name = m[1];
-				seq = new Bytes();
-			}
-		} else seq.set(buf);
-	}
-	if (seq != null && name != null) {
-		seqlen.push([name, seq.length]);
-		h[name] = seq;
+	if (typeof k8_version == "undefined") { // for k8-0.x
+		var seq = null, name = null, gt = '>'.charCodeAt(0);
+		while (file.readline(buf) >= 0) {
+			if (buf[0] == gt) {
+				if (seq != null && name != null) {
+					seqlen.push([name, seq.length]);
+					h[name] = seq;
+					name = seq = null;
+				}
+				var m, line = buf.toString();
+				if ((m = /^>(\S+)/.exec(line)) != null) {
+					name = m[1];
+					seq = new Bytes();
+				}
+			} else seq.set(buf);
+		}
+		if (seq != null && name != null) {
+			seqlen.push([name, seq.length]);
+			h[name] = seq;
+		}
+	} else { // for k8-1.x
+		var seq = null, name = null;
+		while (file.readline(buf) >= 0) {
+			var line = buf.toString();
+			if (line[0] == ">") {
+				if (seq != null && name != null) {
+					seqlen.push([name, seq.length]);
+					h[name] = new Uint8Array(seq.buffer);
+					name = seq = null;
+				}
+				var m;
+				if ((m = /^>(\S+)/.exec(line)) != null) {
+					name = m[1];
+					seq = new Bytes();
+				}
+			} else seq.set(line);
+		}
+		if (seq != null && name != null) {
+			seqlen.push([name, seq.length]);
+			h[name] = new Uint8Array(seq.buffer);
+		}
 	}
 	buf.destroy();
 	file.close();
@@ -161,16 +185,27 @@ function fasta_read(fn)
 
 function fasta_free(fa)
 {
-	for (var name in fa)
-		fa[name].destroy();
+	if (typeof k8_version == "undefined")
+		for (var name in fa)
+			fa[name].destroy();
+	// FIXME: for k8-1.0, sequences are not freed. This is ok for now but not general.
 }
 
 Bytes.prototype.reverse = function()
 {
-	for (var i = 0; i < this.length>>1; ++i) {
-		var tmp = this[i];
-		this[i] = this[this.length - i - 1];
-		this[this.length - i - 1] = tmp;
+	if (typeof k8_version === "undefined") { // k8-0.x
+		for (var i = 0; i < this.length>>1; ++i) {
+			var tmp = this[i];
+			this[i] = this[this.length - i - 1];
+			this[this.length - i - 1] = tmp;
+		}
+	} else { // k8-1.x
+		var buf = new Uint8Array(this.buffer);
+		for (var i = 0; i < buf.length>>1; ++i) {
+			var tmp = buf[i];
+			buf[i] = buf[buf.length - i - 1];
+			buf[buf.length - i - 1] = tmp;
+		}
 	}
 }
 
@@ -185,13 +220,24 @@ Bytes.prototype.revcomp = function()
 		for (var i = 0; i < s1.length; ++i)
 			Bytes.rctab[s1.charCodeAt(i)] = s2.charCodeAt(i);
 	}
-	for (var i = 0; i < this.length>>1; ++i) {
-		var tmp = this[this.length - i - 1];
-		this[this.length - i - 1] = Bytes.rctab[this[i]];
-		this[i] = Bytes.rctab[tmp];
+	if (typeof k8_version === "undefined") { // k8-0.x
+		for (var i = 0; i < this.length>>1; ++i) {
+			var tmp = this[this.length - i - 1];
+			this[this.length - i - 1] = Bytes.rctab[this[i]];
+			this[i] = Bytes.rctab[tmp];
+		}
+		if (this.length&1)
+			this[this.length>>1] = Bytes.rctab[this[this.length>>1]];
+	} else { // k8-1.x
+		var buf = new Uint8Array(this.buffer);
+		for (var i = 0; i < buf.length>>1; ++i) {
+			var tmp = buf[buf.length - i - 1];
+			buf[buf.length - i - 1] = Bytes.rctab[buf[i]];
+			buf[i] = Bytes.rctab[tmp];
+		}
+		if (buf.length&1)
+			buf[buf.length>>1] = Bytes.rctab[buf[buf.length>>1]];
 	}
-	if (this.length&1)
-		this[this.length>>1] = Bytes.rctab[this[this.length>>1]];
 }
 
 /********************
@@ -1694,15 +1740,17 @@ function paf_gff2bed(args)
 
 function paf_sam2paf(args)
 {
-	var c, pri_only = false, long_cs = false;
-	while ((c = getopt(args, "pL")) != null) {
+	var c, pri_only = false, long_cs = false, pri_pri_only = false;
+	while ((c = getopt(args, "pPL")) != null) {
 		if (c == 'p') pri_only = true;
+		else if (c == 'P') pri_pri_only = pri_only = true;
 		else if (c == 'L') long_cs = true;
 	}
 	if (args.length == getopt.ind) {
 		print("Usage: paftools.js sam2paf [options] <in.sam>");
 		print("Options:");
 		print("  -p      convert primary or supplementary alignments only");
+		print("  -P      convert primary alignments only");
 		print("  -L      output the cs tag in the long form");
 		exit(1);
 	}
@@ -1729,6 +1777,7 @@ function paf_sam2paf(args)
 			throw Error("at line " + lineno + ": inconsistent SEQ and QUAL lengths - " + t[9].length + " != " + t[10].length);
 		if (t[2] == '*' || (flag&4) || t[5] == '*') continue;
 		if (pri_only && (flag&0x100)) continue;
+		if (pri_pri_only && (flag&0x900)) continue;
 		var tlen = ctg_len[t[2]];
 		if (tlen == null) throw Error("at line " + lineno + ": can't find the length of contig " + t[2]);
 		// find tags
@@ -1841,7 +1890,10 @@ function paf_sam2paf(args)
 		// optional tags
 		var type = flag&0x100? 'S' : 'P';
 		var tags = ["tp:A:" + type];
-		if (NM != null) tags.push("mm:i:"+mm);
+		if (NM != null) {
+			tags.push("NM:i:"+NM);
+			tags.push("mm:i:"+mm);
+		}
 		tags.push("gn:i:"+(I[1]+D[1]), "go:i:"+(I[0]+D[0]), "cg:Z:" + t[5].replace(/\d+[SH]/g, ''));
 		if (cs_str != null) tags.push("cs:Z:" + cs_str);
 		else if (cs.length > 0) tags.push("cs:Z:" + cs.join(""));
@@ -2051,7 +2103,7 @@ function paf_mapeval(args)
 		warn("Usage: paftools.js mapeval [options] <in.paf>|<in.sam>");
 		warn("Options:");
 		warn("  -r FLOAT   mapping correct if overlap_length/union_length>FLOAT [" + ovlp_ratio + "]");
-		warn("  -Q INT     print wrong mappings with mapQ>INT [don't print]");
+		warn("  -Q INT     print wrong mappings with mapQ>=INT [don't print]");
 		warn("  -m INT     0: eval the longest aln only; 1: first aln only; 2: all primary aln [0]");
 		exit(1);
 	}

data/ext/minimap2/mmpriv.h

@@ -14,6 +14,7 @@
 #define MM_DBG_PRINT_SEED    0x4
 #define MM_DBG_PRINT_ALN_SEQ 0x8
 #define MM_DBG_PRINT_CHAIN   0x10
+#define MM_DBG_SEED_FREQ     0x20
 
 #define MM_SEED_LONG_JOIN  (1ULL<<40)
 #define MM_SEED_IGNORE     (1ULL<<41)
@@ -79,8 +80,6 @@ int mm_idx_getseq2(const mm_idx_t *mi, int is_rev, uint32_t rid, uint32_t st, ui
 mm_reg1_t *mm_align_skeleton(void *km, const mm_mapopt_t *opt, const mm_idx_t *mi, int qlen, const char *qstr, int *n_regs_, mm_reg1_t *regs, mm128_t *a);
 mm_reg1_t *mm_gen_regs(void *km, uint32_t hash, int qlen, int n_u, uint64_t *u, mm128_t *a, int is_qstrand);
 
-mm128_t *mm_chain_dp(int max_dist_x, int max_dist_y, int bw, int max_skip, int max_iter, int min_cnt, int min_sc, float gap_scale,
-					 int is_cdna, int n_segs, int64_t n, mm128_t *a, int *n_u_, uint64_t **_u, void *km);
 mm128_t *mg_lchain_dp(int max_dist_x, int max_dist_y, int bw, int max_skip, int max_iter, int min_cnt, int min_sc, float chn_pen_gap, float chn_pen_skip,
 					  int is_cdna, int n_segs, int64_t n, mm128_t *a, int *n_u_, uint64_t **_u, void *km);
 mm128_t *mg_lchain_rmq(int max_dist, int max_dist_inner, int bw, int max_chn_skip, int cap_rmq_size, int min_cnt, int min_sc, float chn_pen_gap, float chn_pen_skip,

data/ext/minimap2/options.c

@@ -45,6 +45,7 @@ void mm_mapopt_init(mm_mapopt_t *opt)
 	opt->alt_drop = 0.15f;
 
 	opt->a = 2, opt->b = 4, opt->q = 4, opt->e = 2, opt->q2 = 24, opt->e2 = 1;
+	opt->transition = 0;
 	opt->sc_ambi = 1;
 	opt->zdrop = 400, opt->zdrop_inv = 200;
 	opt->end_bonus = -1;
@@ -54,7 +55,7 @@ void mm_mapopt_init(mm_mapopt_t *opt)
 	opt->max_clip_ratio = 1.0f;
 	opt->mini_batch_size = 500000000;
 	opt->max_sw_mat = 100000000;
-	opt->cap_kalloc = 1000000000;
+	opt->cap_kalloc = 500000000;
 
 	opt->rank_min_len = 500;
 	opt->rank_frac = 0.9f;
@@ -90,7 +91,7 @@ int mm_set_opt(const char *preset, mm_idxopt_t *io, mm_mapopt_t *mo)
 	if (preset == 0) {
 		mm_idxopt_init(io);
 		mm_mapopt_init(mo);
-	} else if (strcmp(preset, "map-ont") == 0) { // this is the same as the default
+	} else if (strcmp(preset, "lr") == 0 || strcmp(preset, "map-ont") == 0) { // this is the same as the default
 	} else if (strcmp(preset, "ava-ont") == 0) {
 		io->flag = 0, io->k = 15, io->w = 5;
 		mo->flag |= MM_F_ALL_CHAINS | MM_F_NO_DIAG | MM_F_NO_DUAL | MM_F_NO_LJOIN;
@@ -105,13 +106,30 @@ int mm_set_opt(const char *preset, mm_idxopt_t *io, mm_mapopt_t *mo)
 		mo->min_chain_score = 100, mo->pri_ratio = 0.0f, mo->max_chain_skip = 25;
 		mo->bw_long = mo->bw;
 		mo->occ_dist = 0;
-	} else if (strcmp(preset, "map-hifi") == 0 || strcmp(preset, "map-ccs") == 0) {
+	} else if (strcmp(preset, "lr:hq") == 0 || strcmp(preset, "map-hifi") == 0 || strcmp(preset, "map-ccs") == 0) {
 		io->flag = 0, io->k = 19, io->w = 19;
 		mo->max_gap = 10000;
-		mo->a = 1, mo->b = 4, mo->q = 6, mo->q2 = 26, mo->e = 2, mo->e2 = 1;
-		mo->occ_dist = 500;
 		mo->min_mid_occ = 50, mo->max_mid_occ = 500;
-		mo->min_dp_max = 200;
+		if (strcmp(preset, "map-hifi") == 0 || strcmp(preset, "map-ccs") == 0) {
+			mo->a = 1, mo->b = 4, mo->q = 6, mo->q2 = 26, mo->e = 2, mo->e2 = 1;
+			mo->min_dp_max = 200;
+		}
+	} else if (strcmp(preset, "lr:hqae") == 0) { // high-quality assembly evaluation
+		io->flag = 0, io->k = 25, io->w = 51;
+		mo->flag |= MM_F_RMQ;
+		mo->min_mid_occ = 50, mo->max_mid_occ = 500;
+		mo->rmq_inner_dist = 5000;
+		mo->occ_dist = 200;
+		mo->best_n = 100;
+		mo->chain_gap_scale = 5.0f;
+	} else if (strcmp(preset, "map-iclr-prerender") == 0) {
+		io->flag = 0, io->k = 15;
+		mo->b = 6, mo->transition = 1;
+		mo->q = 10, mo->q2 = 50;
+	} else if (strcmp(preset, "map-iclr") == 0) {
+		io->flag = 0, io->k = 19;
+		mo->b = 6, mo->transition = 4;
+		mo->q = 10, mo->q2 = 50;
 	} else if (strncmp(preset, "asm", 3) == 0) {
 		io->flag = 0, io->k = 19, io->w = 19;
 		mo->bw = 1000, mo->bw_long = 100000;
@@ -156,7 +174,7 @@ int mm_set_opt(const char *preset, mm_idxopt_t *io, mm_mapopt_t *mo)
 		mo->junc_bonus = 9;
 		mo->zdrop = 200, mo->zdrop_inv = 100; // because mo->a is halved
 		if (strcmp(preset, "splice:hq") == 0)
-			mo->junc_bonus = 5, mo->b = 4, mo->q = 6, mo->q2 = 24;
+			mo->noncan = 5, mo->b = 4, mo->q = 6, mo->q2 = 24;
 	} else return -1;
 	return 0;
 }

data/ext/minimap2/python/README.rst

@@ -77,7 +77,9 @@ This constructor accepts the following arguments:
 
 * **min_chain_score**: minimum chaing score
 
-* **bw**: chaining and alignment band width
+* **bw**: chaining and alignment band width (initial chaining and extension)
+
+* **bw_long**: chaining and alignment band width (RMQ-based rechaining and closing gaps)
 
 * **best_n**: max number of alignments to return
 

data/ext/minimap2/python/cmappy.pxd

@@ -36,6 +36,7 @@ cdef extern from "minimap.h":
 		float alt_drop
 
 		int a, b, q, e, q2, e2
+		int transition
 		int sc_ambi
 		int noncan
 		int junc_bonus

data/ext/minimap2/python/mappy.pyx

@@ -3,7 +3,7 @@ from libc.stdlib cimport free
 cimport cmappy
 import sys
 
-__version__ = '2.26'
+__version__ = '2.28'
 
 cmappy.mm_reset_timer()
 
@@ -96,6 +96,7 @@ cdef class Alignment:
 		a = [str(self._q_st), str(self._q_en), strand, self._ctg, str(self._ctg_len), str(self._r_st), str(self._r_en),
 			str(self._mlen), str(self._blen), str(self._mapq), tp, ts, "cg:Z:" + self.cigar_str]
 		if self._cs != "": a.append("cs:Z:" + self._cs)
+		if self._MD != "": a.append("MD:Z:" + self._MD)
 		return "\t".join(a)
 
 cdef class ThreadBuffer:
@@ -112,7 +113,7 @@ cdef class Aligner:
 	cdef cmappy.mm_idxopt_t idx_opt
 	cdef cmappy.mm_mapopt_t map_opt
 
-	def __cinit__(self, fn_idx_in=None, preset=None, k=None, w=None, min_cnt=None, min_chain_score=None, min_dp_score=None, bw=None, best_n=None, n_threads=3, fn_idx_out=None, max_frag_len=None, extra_flags=None, seq=None, scoring=None):
+	def __cinit__(self, fn_idx_in=None, preset=None, k=None, w=None, min_cnt=None, min_chain_score=None, min_dp_score=None, bw=None, bw_long=None, best_n=None, n_threads=3, fn_idx_out=None, max_frag_len=None, extra_flags=None, seq=None, scoring=None):
 		self._idx = NULL
 		cmappy.mm_set_opt(NULL, &self.idx_opt, &self.map_opt) # set the default options
 		if preset is not None:
@@ -125,6 +126,7 @@ cdef class Aligner:
 		if min_chain_score is not None: self.map_opt.min_chain_score = min_chain_score
 		if min_dp_score is not None: self.map_opt.min_dp_max = min_dp_score
 		if bw is not None: self.map_opt.bw = bw
+		if bw_long is not None: self.map_opt.bw_long = bw_long
 		if best_n is not None: self.map_opt.best_n = best_n
 		if max_frag_len is not None: self.map_opt.max_frag_len = max_frag_len
 		if extra_flags is not None: self.map_opt.flag |= extra_flags

data/ext/minimap2/python/minimap2.py

@@ -5,7 +5,7 @@ import getopt
 import mappy as mp
 
 def main(argv):
-	opts, args = getopt.getopt(argv[1:], "x:n:m:k:w:r:c")
+	opts, args = getopt.getopt(argv[1:], "x:n:m:k:w:r:cM")
 	if len(args) < 2:
 		print("Usage: minimap2.py [options] <ref.fa>|<ref.mmi> <query.fq>")
 		print("Options:")
@@ -16,10 +16,11 @@ def main(argv):
 		print("  -w INT      minimizer window length")
 		print("  -r INT      band width")
 		print("  -c          output the cs tag")
+		print("  -M          output the MD tag")
 		sys.exit(1)
 
 	preset = min_cnt = min_sc = k = w = bw = None
-	out_cs = False
+	out_cs = out_MD = False
 	for opt, arg in opts:
 		if opt == '-x': preset = arg
 		elif opt == '-n': min_cnt = int(arg)
@@ -28,11 +29,12 @@ def main(argv):
 		elif opt == '-k': k = int(arg)
 		elif opt == '-w': w = int(arg)
 		elif opt == '-c': out_cs = True
+		elif opt == '-M': out_MD = True
 
 	a = mp.Aligner(args[0], preset=preset, min_cnt=min_cnt, min_chain_score=min_sc, k=k, w=w, bw=bw)
 	if not a: raise Exception("ERROR: failed to load/build index file '{}'".format(args[0]))
 	for name, seq, qual in mp.fastx_read(args[1]): # read one sequence
-		for h in a.map(seq, cs=out_cs): # traverse hits
+		for h in a.map(seq, cs=out_cs, MD=out_MD): # traverse hits
 			print('{}\t{}\t{}'.format(name, len(seq), h))
 
 if __name__ == "__main__":

data/ext/minimap2/seed.c

@@ -112,7 +112,8 @@ mm_seed_t *mm_collect_matches(void *km, int *_n_m, int qlen, int max_occ, int ma
 	}
 	for (i = 0, n_m = 0, *rep_len = 0, *n_a = 0; i < n_m0; ++i) {
 		mm_seed_t *q = &m[i];
-		//fprintf(stderr, "X\t%d\t%d\t%d\n", q->q_pos>>1, q->n, q->flt);
+		if (mm_dbg_flag & MM_DBG_SEED_FREQ)
+			fprintf(stderr, "SF\t%d\t%d\t%d\n", q->q_pos>>1, q->n, q->flt);
 		if (q->flt) {
 			int en = (q->q_pos >> 1) + 1, st = en - q->q_span;
 			if (st > rep_en) {

data/ext/minimap2/setup.py

@@ -23,7 +23,7 @@ def readme():
 
 setup(
 	name = 'mappy',
-	version = '2.26',
+	version = '2.28',
 	url = 'https://github.com/lh3/minimap2',
 	description = 'Minimap2 python binding',
 	long_description = readme(),

data/lib/minimap2/aligner.rb

@@ -21,10 +21,11 @@ module Minimap2
     #   * ava-ont : Nanopore read overlap
     # @param k [Integer] k-mer length, no larger than 28.
     # @param w [Integer] minimizer window size, no larger than 255.
-    # @param min_cnt [Integer] mininum number of minimizers on a chain.
-    # @param min_chain_score [Integer] minimum chaing score.
+    # @param min_cnt [Integer] minimum number of minimizers on a chain.
+    # @param min_chain_score [Integer] minimum chain score.
     # @param min_dp_score
-    # @param bw [Integer] chaining and alignment band width.
+    # @param bw [Integer] chaining and alignment band width. (initial chaining and extension)
+    # @param bw_long [Integer] chaining and alignment band width (RMQ-based rechaining and closing gaps)
     # @param best_n [Integer] max number of alignments to return.
     # @param n_threads [Integer] number of indexing threads.
     # @param fn_idx_out [String] name of file to which the index is written.
@@ -47,6 +48,7 @@ module Minimap2
       min_chain_score: nil,
       min_dp_score: nil,
       bw: nil,
+      bw_long: nil,
       best_n: nil,
       n_threads: 3,
       fn_idx_out: nil,
@@ -72,6 +74,7 @@ module Minimap2
       map_opt[:min_chain_score] = min_chain_score if min_chain_score
       map_opt[:min_dp_max] = min_dp_score if min_dp_score
       map_opt[:bw] = bw if bw
+      map_opt[:bw_long] = bw_long if bw_long
       map_opt[:best_n] = best_n if best_n
       map_opt[:max_frag_len] = max_frag_len if max_frag_len
       map_opt[:flag] |= extra_flags if extra_flags

data/lib/minimap2/alignment.rb

@@ -23,7 +23,7 @@ module Minimap2
   #   @return [Integer] length of the matching bases in the alignment,
   #     excluding ambiguous base matches.
   # @!attribute nm
-  #   @return [Integer] number of mismatches, gaps and ambiguous poistions in the alignment.
+  #   @return [Integer] number of mismatches, gaps and ambiguous positions in the alignment.
   # @!attribute primary
   #   @return [Integer] if the alignment is primary (typically the best and the first to generate)
   # @!attribute q_st
@@ -107,6 +107,7 @@ module Minimap2
       a = [@q_st, @q_en, strand, @ctg, @ctg_len, @r_st, @r_en,
            @mlen, @blen, @mapq, tp, ts, "cg:Z:#{@cigar_str}"]
       a << "cs:Z:#{@cs}" if @cs
+      a << "MD:Z:#{@md}" if @md
       a.join("\t")
     end
   end

data/lib/minimap2/ffi/constants.rb

@@ -40,6 +40,7 @@ module Minimap2
     NO_HASH_NAME     = 0x400000000
     SPLICE_OLD       = 0x800000000
     SECONDARY_SEQ    = 0x1000000000 # output SEQ field for seqondary alignments using hard clipping
+    OUT_DS           = 0x2000000000
 
     HPC              = 0x1
     NO_SEQ           = 0x2
@@ -109,8 +110,10 @@ module Minimap2
         :dp_score,            :int32,   # DP score
         :dp_max,              :int32,   # score of the max-scoring segment
         :dp_max2,             :int32,   # score of the best alternate mappings
+        :dp_max0,             :int32,   # DP score before mm_update_dp_max() adjustment
         :n_ambi_trans_strand, :uint32,
         :n_cigar,             :uint32
+      # :cigar,               :pointer  # variable length array (see cigar method below)
 
       bit_field :n_ambi_trans_strand,
                 :n_ambi, 30,      # number of ambiguous bases
@@ -204,6 +207,7 @@ module Minimap2
         :e,                    :int,     # gap-ext
         :q2,                   :int,     # gap-open
         :e2,                   :int,     # gap-ext
+        :transition,           :int,     # transition mismatch score (A:G, C:T)
         :sc_ambi,              :int,     # score when one or both bases are "N"
         :noncan,               :int,     # cost of non-canonical splicing sites
         :junc_bonus,           :int,
@@ -223,7 +227,7 @@ module Minimap2
         :q_occ_frac,           :float,
         :min_mid_occ,          :int32,
         :max_mid_occ,          :int32,
-        :mid_occ,              :int32, # ignore seeds with occurrences above this threshold
+        :mid_occ,              :int32,   # ignore seeds with occurrences above this threshold
         :max_occ,              :int32,
         :max_max_occ,          :int32,
         :occ_dist,             :int32,

data/lib/minimap2/ffi/functions.rb

@@ -15,10 +15,11 @@ module Minimap2
     private_class_method :mm_set_opt_raw
 
     def self.mm_set_opt(preset, io, mo)
-      ptr = if preset
-              ::FFI::MemoryPointer.from_string(preset.to_s)
-            else
+      ptr = case preset
+            when 0, nil
               ::FFI::Pointer.new(:int, 0)
+            else
+              ::FFI::MemoryPointer.from_string(preset.to_s)
             end
       mm_set_opt_raw(ptr, io, mo)
     end
@@ -77,5 +78,17 @@ module Minimap2
       :mm_gen_md, :mm_gen_MD, # Avoid uppercase letters in method names.
       [:pointer, :pointer, :pointer, Idx.by_ref, Reg1.by_ref, :string],
       :int
+
+    attach_function \
+      :mm_mapopt_init,
+      [MapOpt.by_ref],
+      :void
+
+    # mmpriv.h
+
+    attach_function \
+      :mm_idxopt_init,
+      [IdxOpt.by_ref],
+      :void
   end
 end

data/lib/minimap2/ffi.rb

@@ -2,6 +2,7 @@
 
 # bit fields
 require "ffi/bit_struct"
+
 module Minimap2
   # Native APIs
   module FFI